1. Field of the Invention
The present invention relates to a technique for determining an error or automatically determining a scanning direction in the use of a colorimeter that is manually operated to measure a color chart including an array of color patches on a row-by-row basis.
2. Description of the Related Art
According to one of color chart measuring methods, a user manually positions a colorimeter and manually slides the colorimeter along each row of color patches to perform colorimetry on the color patches, i.e., measure the colors of the color patches on a row-by-row basis. The colorimeter samples colorimetric data at very short time intervals during sliding and analyzes the acquired colorimetric data to automatically recognize each boundary between adjacent patches such that the colorimetric values of the color patches in one row can be obtained by one sliding operation. Such a colorimeter is smaller and more inexpensive than a colorimeter that is automatically positioned. Accordingly, the method using the above-described colorimeter is easily introduced into small-scale firms and small office home offices.
In a colorimetric system using such a colorimeter, a user's operation error causes the system to not properly measure the colors of patches expected. Probable user's operation errors include an error caused by positioning of the colorimeter relative to an improper row of color patches, an error caused by partial measurement of a proper row (i.e., the colorimeter is positioned relative to a proper row but the proper row is partially subjected to colorimetry), and an error caused when sliding is performed in the backward direction. In addition, in some cases, a user's inaccurate sliding operation causes the colorimeter to be shifted from a proper row to the adjacent row during sliding. Furthermore, if the colorimeter misrecognizes the boundary between adjacent patches, accurate colorimetry is not performed. In this case, a patch is skipped or the colorimetric value of a nonexistent patch is added. Upon occurrence of both of the skip and the addition, although the total number of patches is correct, the acquired colorimetric data includes an error (hereinafter, referred to as “colorimetric data error”).
In the use of the manually-operated colorimeter, the following error determination technique has been generally used. Error determination is performed using two values, i.e., a reference value for each patch of a chart to be color-measured and a threshold value serving as a tolerance. A system has previously stored these values and compares a measured value of a patch with the reference value. When the measured value deviates from the reference value by more than the threshold value, the system determines that colorimetry has failed, namely, an error of colorimetry. For example, Japanese Patent Laid-Open No. 2005-061926 discloses such a colorimetric system.
The related-art method, however, has the following disadvantages.
The system has previously stored the reference values for the respective color patches. Accordingly, when the color of a patch deviates from the corresponding reference value by more than a predetermined threshold value, an error is determined even though the color of the patch is correctly measured. For example, to generate a color profile for a printer, a user allows the printer to print a color chart onto a sheet of paper on which the user intends to generate a profile. The user waits until the colors reach stability and then measures the colors in the color chart. Sheets onto which the user intends to generate a color profile vary. In some cases, therefore, a specific color among the colors of patches printed is emphasized depending on the characteristics of a sheet. In the use of a colored sheet, the color of the printed sheet appears in a pale color. In this case, if the threshold value is too strict, an error is determined even through colorimetry is correctly performed. Unfortunately, the user cannot continue colorimetry.
Assuming that the threshold value is loosened to avoid the above-described problem, when an error actually occurs, the error cannot be determined.
As described above, it is difficult to properly determine a reference value and a threshold value for each patch. It is necessary to determine a reference value and a threshold value for each patch and the order of arrangement of color patches by trial and error and assure the quality of the values and the arrangement order. Therefore, it takes long time and high cost. In some cases, error determination is not correctly performed because a reference value and a threshold value are improper.